Learning Science

The Forgetting Curve and Spaced Repetition: Why Kids Forget What They Learn

"Last month they knew this topic cold — then scored zero on the test." Sound familiar? The problem isn't your child; it's how the human brain works. A curve discovered 140 years ago explains it — and there's a proven way to beat it.

Ebbinghaus's discovery: Forgetting is the default

In 1885, German psychologist Hermann Ebbinghaus ran memory experiments on himself and arrived at learning science's most fundamental finding: newly learned information decays exponentially unless it is revisited. The steepest loss happens in the first 24 hours — this model became known as the "forgetting curve." Modern studies show the curve varies by person and material, but the core message holds: forgetting isn't a flaw, it's the brain's default behavior. The brain operates on a simple rule: information it never encounters again must not matter.

Why "we finished that unit" isn't enough

School is linear: teach the topic, test it, move on. But the brain isn't linear. Fractions covered in February, never recalled again until April, simply can't be found in memory on test day. The child forgot not because they "didn't study," but because the knowledge was never recalled at the right time. Night-before cramming seems to work short-term, but it doesn't flatten the curve — the test ends, and the knowledge leaves with it.

The fix: Spaced repetition

One of cognitive science's most strongly evidenced strategies is spaced repetition: recalling the same material at increasing intervals. Every successful recall flattens the forgetting curve a little more; the knowledge survives longer each time. A practical schedule:

The second key: Active recall

Spaced repetition's twin is active recall (retrieval practice). Rereading the notebook feels like reviewing but leaves a weak trace. What actually strengthens knowledge is trying to remember without looking — solving problems, teaching it back, writing what you know on a blank page. The moment of effortful recall is precisely the moment learning happens. So the goal isn't "reviewed it easily" — it's "recalled it with effort."

Why math needs spaced repetition most

• Chained structure: A child who forgets fractions can't learn ratios; forgetting ratios blocks percentages. In math, forgetting costs you forward, not backward. Our exam prep roadmap shows how to work through this chain topic by topic.
• Knowledge plus skill: Math demands both recall (formulas, rules) and skill (application). Both survive only through practice.
• Test structure: School exams and standardized tests cover everything from the start of the term. A student who only knows the latest unit loses most of the test.

How to apply it at home

• Mix the practice: Choose problem sets that include older topics, not just this week's. This is called "interleaving," and it makes a measurable difference on its own.
• The "teach me" game: Have your child teach you the topic. Wherever they get stuck explaining is exactly what needs review.
• Short and frequent wins: 15–20 minutes daily beats 2 hours once a week. The curve loves frequency.
• Normalize forgetting: Ban the phrase "how could you forget?" Forgetting is part of learning; blame only adds anxiety.

How Oyster automates this

The hard part of spaced repetition isn't the knowledge — it's the discipline: which topic, when, with which problem? In Oyster, AI manages that calendar. Weekly planet quests introduce the new topic while problem sets automatically blend in earlier weeks' skills; problems the child got wrong come back more often, mastered ones less often. For parents the outcome is simple: 15–20 minutes of play a day runs a scientific review schedule in the background.